Issues in Science and Technology Librarianship | Summer 1997 |
---|
Journal titles that only offered electronic access did not have a record in our online catalog. URLs were added to existing MARC records of titles with print counterparts in our library. This added yet another point of access that required maintenance by our staff.
Figure 1 shows the increase of electronic journals to which we provide access. Current suppliers include Academic Press, IEE, Springer, IOP, and AIP. We expect access to approximately 300 more titles from Elsevier.
We first decided to limit our selection of electronic journals to full-text or selected full-text titles. Tables of contents with or without abstracts are not included. We feel that our users expect more content in an electronic journal than can be found in a citation database. Electronic journals that deliver only tables of contents with or without abstracts can still be found on appropriate subject web pages.
Determining access points was the next decision. Based on user input, the new model includes an alphabetic listing by journal title and listings by subject category. The ability to search by journal title is also included. Figure 2 is the first page presented to the user.
Having defined the inclusion parameters and the access points, we then studied what information should be presented to the user. We surfed the web to gain ideas from other institutions. We decided to include the following information: journal title, URL, access limitations, passwords, electronic holdings, format, description and subjects. Format is defined as full-text/full-image, selected full-text, etc. We included a LANL specific subject code to assist in collection development.
A two-layer approach is used in displaying this information. The first level pages (alphabetic or subject) display the journal title, format (if other than full-text), electronic holdings, password link (where appropriate) and a description link. An icon is displayed to indicate access restrictions. The journal titles serves as the active hyperlink. If there are multiple URLs for a single title, the journal title is not hyperlinked; each URL is displayed with its own holdings statement. Each URL has a unique description page. Below is the beginning of the physics subject page.
The second level description page presents all of the information to the user. Passwords are displayed only to users with valid LANL IP addresses. The description paragraph details subject coverage and presentation of the journal. Subjects list the categories assigned to that journal. Below is the description page for the title Applied physics letters.
At this point, we created several mock-ups of the different pages. We gathered input from our users and library staff. Changes were incorporated from this feedback.
Cataloging decisions were the next step. We reviewed pros and cons of separate MARC records for each format or one MARC record per title. Our current process is to use one MARC record per title. We agree this is the best method for the end user and continue to have one MARC record per title. Next we reviewed what information we wanted to present to the user to determine where this information would reside in the MARC record. We created a 956 local tag to accommodate all of the information. The 956 is table driven which minimizes input and delivers consistent results. The information in the 956 does not display in the online catalog. Below is a table of the information and MARC tag where the information resides.
Information | MARC tag | |
---|---|---|
Journal title | 245 | |
ISSN | 022 | |
URL | 856u | |
Electronic holdings | Holdings - copy set | |
Format | 956a | |
Subjects | 956b | |
Access limitation | 956d | |
Password | 956e | |
Description | 956f |
To promote consistency of information provided to the cataloger, we devised an electronic journal checklist. This checklist contains the journal information needed to create the MARC record and consequently the web pages. Subject specialists are responsible for selecting electronic journals and completing the checklists. Each new title is discussed in a selection meeting to ensure that all relevant subject categories are assigned and to alert subject specialists to new titles.
On the webserver, a script converts the text file into a set of HTML pages. A page containing an alphabetical list of all titles is created. Separate pages for each subject category are also created. The description pages are dynamically created at the time of request. A key feature of the HTML pages is that they are "force-loaded"; meaning that the web browser must load the page each time it is accessed, rather than using its cache. This way, we are confident that users are seeing the latest information.
A CGI script on the web server also uses the text file for title word searches. Results of each search are converted to HTML on the fly. The conversion and search scripts are written in Perl. The search script uses Perl's built-in pattern matching function.
A web robot validates the URLs on a monthly basis. The journal cataloger receives a report on what URLs are not valid.
Have we maintained multiple access points? Yes, an electronic journal can have multiple subject categories. An added benefit occurred with the introduction of our web catalog in that all electronic journals are present with active links. Access points have actually increased.
Have consistency and timeliness improved? With one place to update or change a URL, the answer has to be yes. Timeliness is still questionable, as only one cataloger is responsible for electronic journals. Some form of backup needs to be provided.
What about minimizing staff time? The initial roll out of the new model consumed a great deal of staff time and energy. This was expected. Academic Press has given us an opportunity to test time savings of our new model. We have access to 100 Academic Press titles. These 100 titles have been assigned 114 subject categories. In the old model, five people would have been involved, changing 214 URLs on web pages plus one person changing 100 URLs in the online catalog. In the new model, one person was involved changing 100 URL's in the online catalog.
We consider the new model a success. Our major goals have been achieved. We continue to investigate other web pages that could benefit from this model. Our electronic database page is currently being revamped. We also continue to explore new ways to use MARC records and our online catalog.
We welcome your comments about this article. Please fill out this form for possible inclusion in a future issue.